Vouchers, Tax Credits, ESAs and Charters Compared for State Lawmakers

Lawmakers love to say they are for "school choice," the way a restaurant says it is for "food." The phrase covers a lot of dishes, and some of them are going to get sent back to the kitchen by a state supreme court. If you are going to draft a bill, testify on one or vote on one, you need to know which dish you are ordering, what it costs and what the judges in your state have already said about it.

This is the compare-and-contrast edition. Four mechanisms, one table, then a plain word on who each one suits and where each one runs into a state constitution. If you want the explainer version for parents, see Vouchers, ESAs, Tax Credits and Charters Sorted Out in One Table. This one is for the person who has to write the bill.

The table

Now, the details.

Vouchers: the original

A voucher is a certificate, paid for by the state, that a family can take to a participating private school. In the standard version, the state pays the school, either directly or through the family, a set amount per student.

The modern history begins in Milwaukee. The Milwaukee Parental Choice Program started in 1990, the first modern, publicly funded voucher program in the country. It was originally small and limited to non-religious private schools. In 1995 Wisconsin extended it to religious schools, and in 1998, in Jackson v. Benson, the Wisconsin Supreme Court upheld the expanded program. I say "the modern history" because it was a bipartisan project: its champions included a Democratic legislator from Milwaukee who wanted more options for poor families, along with a Republican governor. The usual caricature does not fit.

The federal landmark is Zelman v. Simmons-Harris in 2002. Ohio had a voucher program in Cleveland, and most participating schools were religious. The Supreme Court held that it did not violate the Establishment Clause of the First Amendment, because the program was neutral toward religion and the aid reached schools only through the independent choices of parents. In plain English, if the money follows a genuine private choice, the federal Constitution does not forbid it. For the how-it-works version, see How a School Voucher Works, From the State Check to the Classroom Door.

The weakness: because a voucher is a direct appropriation to families for private schooling, it is the form most exposed to state constitutional challenge. Zelman answered the federal question. State constitutions have their own provisions.

Tax-credit scholarships: the indirect route

A tax-credit scholarship program works differently. Donors give money to a scholarship-granting organization, and the state gives the donor a credit against their state taxes. The organization awards scholarships to children. The state never appropriates the funds; it forgoes revenue it would otherwise have collected.

Arizona was an early adopter, with a credit for donations to school tuition organizations in 1997. The Arizona Supreme Court upheld it under the state constitution in 1999, in Kotterman v. Killian. When taxpayers challenged the program in federal court, the Supreme Court dismissed the case in 2011 in Arizona Christian School Tuition Organization v. Winn, on the ground that the plaintiffs lacked standing to sue as taxpayers.

Why would a lawmaker pick this route? Legally, because many courts treat a tax credit as the taxpayer's own money rather than public money, which can sidestep the constitutional clauses that bar appropriations to religious schools. Politically, because it brings in private donors. Fiscally, because the program is usually capped, which keeps the cost predictable.

The weakness: it is more complicated to administer, depends on donations, and can be more exposed if a court concludes that the credit is really a disguised appropriation. It also depends on the donor's willingness to participate.

Education savings accounts: the flexible version

An ESA takes the idea a step further. Instead of paying a school, the state deposits money into an account the parent controls. The parent can spend it on approved expenses: private-school tuition, tutoring, therapy, textbooks, curriculum and sometimes more. Funds that are not used may in some designs be saved for later years or college.

Arizona created the first ESA program in 2011, initially for children with disabilities. The design appeals to families because a child's needs are rarely met by one institution. A student with a learning difference might attend a small school part of the day and see a therapist, and an ESA can pay for both.

The fiscal and legal issues are slightly different from those of a voucher. Because the money goes to the family, many supporters argue that it is even further removed from a direct payment to a school. On the other side, ESAs raise their own questions: what counts as a permitted expense, how the state audits spending, whether the account can be used for a wide range of products and whether the program can be capped.

The weakness: accountability. A state needs a plain, enforceable rule about what the money may buy, and a process to catch misuse. A program that cannot answer "where did the money go?" is a gift to its critics.

Charter schools: choice inside the public system

A charter school is a public school, funded with public money, but run independently under a contract, the charter, with an authorizer such as a state board, a university or a district. In exchange for flexibility on curriculum, staffing and schedule, the school agrees to meet performance goals, and the authorizer can close it if it fails.

Minnesota passed the first charter law in 1991, an event covered in The 1991 Minnesota Law That Gave America the Charter School. Other states followed within a few years.

Charter schools face fewer religious-school constitutional problems than the others, since they are public and secular. Their legal challenges tend to center on who may authorize a charter, how the funding formula works, and whether local school boards have exclusive control. In some states, courts have debated whether independent charters fit the constitution's language about the public school system.

The weakness: facilities. A charter often receives per-pupil operating funds but no building. Whether the state helps with facilities is a key drafting question. And quality control matters: a charter law with weak authorizers produces weak schools, which gives the whole movement a bad name.

Where each one runs into a state constitution

This is the part that decides the outcome, so here it is without ornament.

Blaine amendments. Many state constitutions contain provisions, often called Blaine amendments, that bar public money from going to religious schools. They have a history I examine in The Blaine Amendments and the Century-Old Fight Over Religious Schools. They were the main barrier for vouchers and tax-credit programs for generations.

What the Supreme Court has said lately. The federal courts have narrowed the force of those provisions. In Espinoza v. Montana Department of Revenue in 2020, the Court held that a state which creates a tax-credit scholarship program may not exclude religious schools solely because they are religious. The Montana program had been struck down by the state's supreme court under the no-aid provision, and the U.S. Supreme Court held that applying the provision that way violated the Free Exercise Clause. In Carson v. Makin in 2022, the Court applied the same principle to a Maine tuition program. Earlier, in Locke v. Davey in 2004, the Court had allowed a state to exclude devotional theology degrees from a scholarship, so the line is not without limits. The practical point for a drafter: if your state creates a program and opens it to private schools, you generally cannot carve out religious ones for being religious.

Uniformity and "general and uniform" clauses. Some state constitutions require a "uniform" system of public schools. In 2006, the Florida Supreme Court in Bush v. Holmes struck down a state voucher program under such a clause. In 2004 the Colorado Supreme Court struck down its pilot program in a case called Owens v. Colorado Congress of Parents, on the ground that it violated the local control of instruction provision. That is a reason to read your own state's text and cases before you choose a vehicle.

Appropriations clauses and public-purpose rules. Some states restrict spending on non-public institutions. These matter most for vouchers and ESAs.

A quick fiscal sketch

Numbers make the differences concrete, so here is a labelled hypothetical with made-up figures in a made-up state. Suppose the state spends $9,000 per pupil on public schools.

• Voucher: the legislature sets the voucher at two-thirds of that figure. Two-thirds of $9,000 is $6,000 (9,000 divided by 3 is 3,000, times 2). If 1,000 students use vouchers, the appropriation is 1,000 times $6,000, or $6,000,000. The cost is visible in the budget, and it grows if enrollment grows unless the legislature caps it.
• Tax-credit scholarship: the legislature caps the total credits at $5,000,000 a year. If the average scholarship is $5,000, the program funds at most 1,000 scholarships (5,000,000 divided by 5,000). The state forgoes at most $5,000,000 in revenue, and the cap is the lever that controls the cost.
• ESA: the legislature deposits $6,000 per participating child in an account. For 1,000 children, that is $6,000,000, the same as the voucher, but the family decides how to spend it.

Notice what the arithmetic does and does not show. It shows that each design has a different budget signature: an open-ended appropriation, a hard cap on forgone revenue, or a per-child deposit. It does not show whether the state saves money or loses it overall, because that depends on how many of those students would otherwise have been in public schools and on which costs actually fall when a student leaves. Anyone who gives you a confident answer without those facts is guessing, and the honest drafter says so in the fiscal note.

Who each one suits

A short guide, for drafting.

• Voucher: best for a state whose constitution and court history are friendly to direct aid, where the goal is a simple, large and predictable program. A means test can focus it on low-income families.
• Tax-credit scholarship: best where the state constitution is hostile to direct aid, or where a legislature prefers a capped, donor-driven program. It tends to be the safer legal path.
• ESA: best where lawmakers want to reach families with special circumstances: a child with disabilities, a military family, a rural child whose nearest school is far away. It demands strong spending rules.
• Charter: best where the goal is more options inside public education and where the political ground is not ready for private-school aid. It demands strong authorizing, and a plan for buildings.

None of these is a trophy. Each is a tool, and a lawmaker should pick it the way a carpenter does: by looking at the job and not the label.

Questions to ask before you draft

1. What does my state constitution say about aid to religious or private schools? Read the text and the cases.
2. What has my state supreme court said about school-choice programs? Look for earlier rulings and for the reasoning, not just the result.
3. Who is eligible? Everyone, or families by income, disability or the performance of the local school? The answer drives the cost and the politics.
4. What does the money follow? The child, the school or the donor?
5. How will the state measure results and catch fraud? Accountability should be plain and enforceable. Parents are a form of accountability, but they cannot audit a ledger.
6. What does it cost, and who pays? Ask your fiscal office for an estimate and review its assumptions. For the debate over costs, see Vouchers Drain the Public Schools, and the Answer Is in the Arithmetic.

The bottom line

• School choice is four different tools: vouchers, tax-credit scholarships, ESAs and charters. They differ in who writes the check, who chooses and which part of the state constitution they bump into.
• Zelman (2002) settled the federal Establishment Clause question for neutral programs, and Espinoza (2020) and Carson (2022) bar a state from excluding religious schools just for being religious. The state-law fights, such as uniformity clauses and local control, remain.
• Choose the vehicle that fits your state's text and court record. In a hostile constitution, a capped tax credit or a charter law is often the safer legal route.
• Build in accountability: a clear rule on allowed spending, audits and a way to close failing schools. The first scandal is the best argument the opposition will ever get.
• Before you call your legislator, read your state constitution's education article and the state supreme court's school-choice decisions.